membranes
Review
Membrane Technology for the Recovery of Lignin:
A Review
Daniel Humpert 1 , Mehrdad Ebrahimi 1 and Peter Czermak 1,2,3, *
1
2
3
*
Institute of Bioprocess Engineering and Pharmaceutical Technology, University of Applied Sciences
Mittelhessen, Giessen 35390, Germany; [email protected] (D.H.);
[email protected] (M.E.)
Department of Chemical Engineering, Kansas State University, Manhattan, KS 66506, USA
Faculty of Biology and Chemistry, Justus-Liebig University of Giessen, Giessen 35390, Germany
Correspondence: [email protected]; Tel.: +49-641-309-2551
Academic Editor: Clàudia Fontàs
Received: 5 July 2016; Accepted: 29 August 2016; Published: 6 September 2016
Abstract: Utilization of renewable resources is becoming increasingly important, and only sustainable
processes that convert such resources into useful products can achieve environmentally beneficial
economic growth. Wastewater from the pulp and paper industry is an unutilized resource offering
the potential to recover valuable products such as lignin, pigments, and water [1]. The recovery of
lignin is particularly important because it has many applications, and membrane technology has
been investigated as the basis of innovative recovery solutions. The concentration of lignin can
be increased from 62 to 285 g·L−1 using membranes and the recovered lignin is extremely pure.
Membrane technology is also scalable and adaptable to different waste liquors from the pulp and
paper industry.
Keywords: ceramic membrane; lignin treatment; lignin fractionation; lignosulfonate; spent sulfite liquor
1. Introduction
Lignin is a phenolic macromolecule that combines with cellulose to form lignocellulose, the most
abundant organic polymer on earth, representing ~30% of the total organic carbon in the biosphere [2].
Lignin accumulates naturally in plant cell walls, a process known as lignification. During pulp
production, lignin must be separated from cellulose because it causes paper products to turn yellow
over time [3]. Lignin can be separated from cellulose by applying mechanical force, or by chemical
modification to make the lignin more soluble, and two major chemical methods have been described:
the kraft process and the sulfite process [4]. The waste material from these processes (black liquor and
spent sulfite liquor, respectively) is a complex mixture of organic and inorganic molecules. The global
paper industry produces more than 5 × 107 tonnes of modified lignin every year, dissolved in the
waste liquor [5]. Approximately 99% of this waste liquor is incinerated to produce energy, and the
remainder is used to produce low-value products such as wood glue and wetting agents. However,
lignin is a versatile raw material that can be used to manufacture many valuable products, including
vanillin, vanillic acid, synthetic tannins and polymer filters [6,7].
2. Characteristics of Lignin Solutions
2.1. Black Liquor
Black liquor is a highly alkaline (pH 13–14) [8] and viscous liquid by-product of the kraft process.
The process was invented in 1879 by the German chemist Carl Ferdinand Dahl in Gdansk [9]. The high
lignin content (kraft lignin) gives the black liquor its characteristic dark brown to black color [10]. Black
Membranes 2016, 6, 42; doi:10.3390/membranes6030042
www.mdpi.com/journal/membranes
Membranes 2016, 6, 42
2 of 13
liquor is rich in organic molecules and inorganic pulping chemicals [11]. The exact composition varies
according to the pulping method and the properties of the wood. Typical black liquor components
are listed in Table 1. The average molecular mass (Mw ) of kraft lignin varies between 1100 and 6500
g·moL−1 , which is lower than soda-anthraquinone lignin, organosolv lignin and ethanol process lignin.
The polydispersity (Mw /Mn ) ranges from 1.8 to 3.6 [12–17]. Furthermore, elementary analysis shows
that lignin from black liquor has a higher carbon content than the other types of lignin listed above,
probably due to dehydration during the pulping process [12]. An important property that determines
the reactivity (Mannich reactivity) of lignin is the number of phenolic hydroxyl groups. Kraft lignin
and soda-anthraquinone lignin have the highest ratio of phenolic hydroxyl groups among the types of
lignin listed above [12].
Table 1. Typical composition of black liquor [8].
Component
Variation Range
dry weight
polyaromatic components
saccharinic acid
formic acid
acetic acid
extractives
methanol
inorganic elements (mainly sodium)
lignin
(12–18) wt %
(30–45) wt %
(25–35) wt %
(0–10) wt %
(0–10) wt %
(3–5) wt %
1 wt %
(17–20) wt %
(45–65) g·L−1
2.2. Spent Sulfite Liquors
Spent sulfite liquor is a highly acidic (pH ~2) liquid by-product of the sulfite process. Like black
liquor, spent sulfite liquor has a dark brown color, as well as a high content of organic molecules and
inorganic pulping chemicals such as sulfurous acid and alkali salts [9]. It also contains large quantities
of various monosaccharides derived from degraded hemicellulose, although the precise composition
depends on the wood source. Typical spent sulfite liquor components are listed in Table 2. Spent
sulfite liquor is generally burned for energy production (the calorific value of lignin is 26,255 kJ·kg−1
compared to 41,840 kJ·kg−1 for oil [18]) and the pulping chemicals are regenerated in a concentration
step [19].
Table 2. Typical composition of spent sulfite liquors before evaporation [19–25].
Component
Variation Range
dry weight
acetic acid
extractives
lignosulfonate
pH value
arabinose
xylose
mannose
galactose
glucose
fucose
rhamnose
furfural
hydroxymethylfurfural
ash
density
methanol
(128–220) g·L−1
(4.7–9.3) g·L−1
~1 g·L−1
(59–120) g·L−1
(1.7–3.4) g·L−1
(1.0–7.8) g·L−1
(0.8–26.7) g·L−1
(4.0–16.16) g·L−1
(0.2–5.34) g·L−1
(1.7–3.28) g·L−1
0.4 g·L−1
~1 g·L−1
(0.03–2.00) g·L−1
~0.34 g·L−1
(19.8–20.8) g·L−1
(1180–1050) g·L−1
<1 g·L−1
Membranes 2016, 6, 42
3 of 13
Relatively little is known about the molecular weight distribution of lignosulfonate molecules
in spent sulfite liquor. This has been investigated by size-exclusion chromatography when the raw
material for the liquor was beech wood, revealing an average molecular mass (Mw ) of 5700 g·moL−1
and a polydispersity (Mw /Mn ) of 5.8 [26]. Another study revealed that the molecular weight
distribution of lignosulfonate in 29 spent sulfite liquor samples was 1030 g·moL−1 and did not
depend on the pulping conditions [27].
3. Technologies for Lignin Purification
The waste liquor from the pulp industry is contaminated with pulping chemicals such as
sodium hydroxide and sodium sulfide [6]. There are currently no cost-efficient purification strategies,
so only a small quantity of valuable lignin products are manufactured. Several methods have
been developed to recover lignin from solution, and the current, state-of-the-art solution involves
a combination of precipitation and membrane filtration, such as microfiltration (MF), ultrafiltration
(UF) and nanofiltration (NF) [28,29]. In the first step, the solubility of lignin is reduced by adding
chemicals such as carbon dioxide, sulfuric acid, or chlorine dioxide, and the precipitated lignin is then
separated by membrane filtration. Flux decline is one of the greatest challenges during filtration [6]
and is difficult to prevent because of the wide molecular weight distribution of the lignin molecules
(0.1–400 kDa) [15,26]. Membrane technology is a promising method to achieve efficient lignin recovery,
and numerous studies have been carried out to investigate the ability of membranes to recover lignin
precipitated from solution (Table 3). Current approaches for the recovery of lignin using membranes
are summarized in Figure 1. Ceramic membranes in particular are ideal for the treatment of spent
sulfite and black liquors because of their ability to tolerate extreme pH values.
Table 3. Overview of recent studies investigating membrane technologies for the treatment of black
liquor and spent sulfite liquor from pulp and paper mills.
Process
Source
Membrane Type
Membrane Material
Reference
UF
UF
UF
UF
UF
UF
UF
NF/UF
MF/UF
UF/NF
UF
UF/NF
–
UF/NF
UF
UF
UF
UF
UF
UF
MF/UF/NF
UF
UF/NF
black liquor
black liquor
black liquor
black liquor
black liquor
black liquor
black liquor
black liquor
black liquor
black liquor
black liquor
black liquor
black liquor
black liquor
black liquor
black liquor
black liquor
black liquor
spent sulfite liquor
spent sulfite liquor
spent sulfite liquor
spent sulfite liquor
spent sulfite liquor
–
flat membranes
tubular membranes
flat membranes
tubular membranes
flat membranes
tubular membranes
tubular membranes
tubular membranes/flat membranes
tubular membranes
flat membranes
flat membranes
flat membranes
tubular membranes
tubular membranes
tubular membranes
flat membranes
tubular membranes
flat membranes
flat membranes
tubular membranes
flat membranes
flat membranes
polymer
polymer
polymer
polymer
ceramic
polymer
ceramic
ceramic
polymer/ceramic
ceramic
polymer
polymer
–
ceramic
ceramic
ceramic
polymer
polymer/ceramic
polymer
polymer
ceramic
polymer
polymer
[18]
[30,31]
[8]
[32]
[33–35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
Membranes 2016, 6, 42
Membranes 2016, 6, 42
4 of 13
4 of 12
Figure 1. Summary
Summary of
of different
different strategies
strategies in
in which
which membrane
membrane fractionation
fractionation can
can be
be used
used for
for the
treatment of waste liquor from the pulp and paper industry. During
During the
the pulping
pulping process,
process, the
the raw
raw
treatment
and
lignin.
TheThe
pulp
can can
be treated
by membrane
filtration
to isolate
material is
is separated
separatedinto
intocellulose
cellulose
and
lignin.
pulp
be treated
by membrane
filtration
to
lignin
molecules.
The
remaining
solution
can
be
recycled
to
the
paper
production.
Additionally,
the
isolate lignin molecules. The remaining solution can be recycled to the paper production.
waste liquor from
the paper
mill
canthe
be paper
treatedmill
by membrane
filtration
to isolatefiltration
the containing
lignin.
Additionally,
the waste
liquor
from
can be treated
by membrane
to isolate
the
The remaining
solution
can be recycled
forcan
energy
generation.
containing
lignin.
The remaining
solution
be recycled
for energy generation.
3.1. Membrane
Treatment of
of Black
Black Liquor
Liquor
Membrane Technology for the Treatment
One of the first examples of lignin
lignin recovery
recovery by
by membrane
membrane filtration
filtration was
was published
published in
in 1988
1988 [18].
[18].
The authors investigated
investigated the chemical
chemical reactivity
reactivity of lignin as a co-polymer during the production of
phenolic plastics,
plastics,asas
well
as separation
the separation
of high-molecular-weight
lignin
a 10 kDa
well
as the
of high-molecular-weight
lignin using
a 10using
kDa polysulfone
polysulfone
membrane.
They
were
able
to
increase
the
lignin
concentration
in
the
retentate
by60.69
~174%
membrane. They were able to increase the lignin concentration in the retentate by ~174% (from
to
−1
−
1
(from
g∙L )aby
applying a permeate/concentrate
ratio of four
in a lignin
166.4 g60.69
·L )toby166.4
applying
permeate/concentrate
ratio of four resulting
in aresulting
lignin purity
of 63%purity
in the
of
63% in theand
concentrate
17% in the
permeate.
After
membrane
they
precipitated
the
concentrate
17% in theand
permeate.
After
membrane
filtration,
theyfiltration,
precipitated
the
lignin fraction
lignin
fraction
with
sulfuric
acid
at
pH
2.
The
purified
lignin
performed
better
as
a
co-polymer
with sulfuric acid at pH 2. The purified lignin performed better as a co-polymer for the productionfor
of
the
production
of plastics than chemically-modified
lignin.
plastics
than chemically-modified
lignin.
Later studies
studiescompared
compared
diafiltration
andprecipitation
acid precipitation
of black
liquor
[30,31].
the the
diafiltration
and acid
of black liquor
[30,31].
Diafiltration
2
2
Diafiltration
achieved
using a plate-and-frame
and 0.36
m polymer
with
was achievedwas
using
a plate-and-frame
module and 0.36module
m polymer
membranes
with membranes
molecular weight
◦ Cmaintained
molecular
weightvalues
cut-offranging
(MWCO)
values
ranging
from
6 toliquor
50 kDa.
The
black liquor
was
cut-off (MWCO)
from
6 to 50
kDa. The
black
was
maintained
at 60
during the
at
60 °C during
the authors
diafiltration
step.90%
The
authors
separated 90% components
of all membrane-permeable
diafiltration
step. The
separated
of all
membrane-permeable
of the liquor with
−2 ·h−1 with
components
of ratio
the liquor
liquor/water
ratio ofmembrane
2:3. They achieved
membrane
a liquor/water
of 2:3.with
Theyaachieved
an average
flux of 90 an
L·maverage
the 25 flux
kDa
of
90 L∙m−2and
∙h−1 with
the 25
kDa
membrane
and concentration,
recovered 54%
of the acid
initial
lignin concentration,
membrane
recovered
54%
of the
initial lignin
whereas
precipitation
achieved a
whereas
acid precipitation
a lignin
recovery
95%.
diafiltration
of polysulfone
black liquor and
has
lignin recovery
of 95%. Theachieved
diafiltration
of black
liquorofhas
alsoThe
been
tested using
also
been tested using
polysulfone
andMWCO
polyethersulfone
membranes
with
MWCO
values
of 4,
8
polyethersulfone
UF membranes
with
values of 4,UF
8 and
20 kDa [8].
Filtration
was
carried
out
−1,
◦
−
1
and
20
kDa
[8].
Filtration
was
carried
out
at
a
temperature
of
60
°C,
a
cross-flow
velocity
of
4
m∙s
at a temperature of 60 C, a cross-flow velocity of 4 m·s , and a transmembrane pressure (TMP) of
and
a transmembrane
pressure
(TMP)
of the
1.0–7.0
bar.
The the
authors
retained
45% of67%
the lignin
using
the
1.0–7.0
bar. The authors
retained
45% of
lignin
using
20 kDa
membrane,
using the
8 kDa
20
kDa membrane,
kDa membrane
80% using
theof41:1
kDa
membrane.
a
membrane
and 80% 67%
usingusing
the 4 the
kDa8membrane.
With aand
feed/water
ratio
they
achieved aWith
lignin
feed/water
ratio
of
1:1
they
achieved
a
lignin
purity
of
78%,
but
the
flux
declined
by
94%.
They
purity of 78%, but the flux declined by 94%. They concluded that membranes with a higher MWCO
concluded
that
membranes
with a higher
MWCO
increase
the purity
ofathe
recovered
whereas
increase the
purity
of the recovered
lignin,
whereas
membranes
with
low
MWCO lignin,
are better
if the
membranes
with a high
low MWCO
are better if the aim is to achieve a high lignin concentration.
aim is to achieve
lignin concentration.
A further
furtherstudy
study
compared
the ability
of UF
and selective
precipitation
the
compared
the ability
of UF and
selective
precipitation
to completetothecomplete
fractionation
fractionation
of the lignin
the
liquor process
of the pulping
For ceramic
the UF,
of the lignin resulting
fromresulting
the blackfrom
liquor
ofblack
the pulping
[46]. Forprocess
the UF,[46].
tubular
tubular
ceramic
membranes
with
a
MWCO
value
of
5,
10,
and
15
kDa
were
investigated,
and
the
membranes with a MWCO value of 5, 10, and 15 kDa were investigated, and the lignin was fractionated
lignin
was fractionated
by the
successively
increasing
the MWCO
values. The
selective
was
by successively
increasing
MWCO values.
The selective
precipitation
was
carried precipitation
out with different
carried out with different concentrations of sulfuric acid. Analytical results obtained with size
exclusion chromatography showed that precipitation as well as UF with ceramic membranes are
Membranes 2016, 6, 42
5 of 13
concentrations of sulfuric acid. Analytical results obtained with size exclusion chromatography showed
that precipitation as well as UF with ceramic membranes are effective techniques to fractionate lignin
in black liquor. For the UF, the polydispersity and the average molar mass of the lignin fractions
decrease with the MWCO value of the ceramic membrane. Fourier transform infrared spectroscopy,
thermal analysis, and proton nuclear magnetic resonance spectroscopy (H-NMR) showed that lignin
from UF contains less contamination with hemicellulose as precipitated lignin fractions. Additionally,
the H-NMR spectra showed that UF slightly depolymerized lignin.
Padilla et al. [47] investigated the ultrafiltrated lignin fractions obtained from industrial black
liquor after alkaline digestion of tropical woods as stabilizers of liquid-liquid and solid-liquid interfaces.
The UF was carried out with diluted black liquor (15% solid content). The diluted black liquor was
prefiltered with a 100 kDa polysulfone membrane. The permeate of this filtration step was used as
a feed for the next filtration process with the 30 kDa polysulfone membrane. This procedure was
repeated for a 10 kDa polysulfone and a 5 kDa and 1 kDa cellulose acetate membrane. It was observed
that an increase of the molecular weight of the lignin fractions resulted in a higher surface activity.
Smaller lignin molecules have an unbalanced ratio of hydrophobic and hydrophilic groups. Thereby
the hydrophilic becomes more dominant and the surface activity is reduced. It follows from the
foregoing that the high molecular weight lignin fraction has the highest dispersion abilities. This was
demonstrated by viscosity measurements of bentonite mixed with high molecular lignin. Furthermore,
it has been shown that only the high molecular weight fraction of lignin has a measurable impact on
the viscosity of a bentonite-lignin solution.
A further study also investigated the behavior of fractionated and ultrafiltrated black liquor
relating to the glass transition temperature and the chemical composition. The cross flow ultrafiltration
was carried out with aluminum oxide, titanium oxide, and regenerated cellulose membranes with
MWCO values of 1, 5, and 10 kDa. The feed had a temperature of 40–65 ◦ C during the ultrafiltration
and the TMP was ~3.5 bar. Overall, six different lignin fractions have been isolated by membrane
filtration (0–1 kDa, 0–5 kDa, 1–5 kDa, 5–10 kDa, >5 kDa, >10 kDa). In the first step, the black liquor
was filtered with the 5 kDa membrane. The permeate of this filtration was additionally fractionated
with the 1 kDa membrane. The retentate of the first step was fractionated with a 10 kD membrane.
Retentate and permeate of both filtration steps were stored. The glass transition temperature was
strongly dependent on the molecular weight of the lignin fraction and varied from 70 to 170 ◦ C with
a rising molecular weight from the 0–1 kDa fraction to the >10 kDa fraction. This relation has been
demonstrated earlier by solvent fractionated lignin. The increasing molecular weight of the lignin
resulted in an increasing entanglement and rigidity of the lignin molecule. That leads, finally, to an
increase in the glass transition temperature. The analysis of the chemical structure of the lignin fraction
was carried out by phosphor nuclear magnetic resonance spectroscopy (P-NMR). It was shown that
the number of phenolic hydroxyl groups decrease whereas the number of aliphatic hydroxyl groups
increase when the molecular size of the lignin fraction increased. The carboxylic groups were not
affected by the molecular weight of the black liquor fraction. This behavior is in accordance with the
mechanism of lignin degradation during the cocking process. In summary, this study showed that
UF of black liquor allows the selective extraction of lignin fractions with particular termo-mechanical
properties [48].
Dead-end and cross-flow filtration using various cellulose acetate UF and NF membranes have
been tested for the treatment of black liquor, to investigate the impact of process and membrane
parameters such as the MWCO (1, 5, and 10 kDa), Reynolds number (3–20,000), TMP (2.75–8.27 bar)
and feed concentration [32]. The authors found that a higher MWCO increases the flux but reduces the
lignin retention efficiency, and that there was a relationship between the Reynolds number and the
flux. The optimal Reynolds number was 17,500 for the radial stirred cell dead-end system, 1200 for the
rectangular stirred cell dead-end system, and four for the radial cross-flow system. A higher Reynolds
number resulted in slower surface layer formation and facilitated backward diffusion. This confirmed
that convection transport is the predominant membrane transport mechanism at high Reynolds
Membranes 2016, 6, 42
6 of 13
numbers based on the low osmotic pressure, resulting in a higher membrane flux. The opposite effect
was caused by irreversible membrane fouling, leading to a slower flux increase at high Reynolds
numbers and, finally, to an asymptotic approximation of a maximum permeate flux. The authors found
that lignin retention increased with higher Reynolds numbers and ran asymptotically to a maximum
value. Furthermore, they recognized that concentration polarization increases at high pressures and
feed concentrations, in agreement with other studies [30,31,36]. The flux decline during filtration was,
therefore, shown to be controlled by the osmotic pressure during cross-flow filtration and by surface
layer formation during dead-end filtration.
The ultrafiltration of black liquor has also been carried out using a 15 kDa aluminum
oxide/titanium oxide membrane with seven channels (diameter = 6 mm) [33]. The lignin retention
was 30%–40% depending on the feed temperature (60 ◦ C, 75 ◦ C, or 90 ◦ C). The crossflow velocity
was kept constant at 4.5 m·s−1 , causing a pressure drop of 0.9 bar. The retention declined at higher
temperatures because the lignin became more soluble, but the retention of the inorganic components
varied depending on their valency. The retention of monovalent ions such as sodium and potassium
was nearly zero, whereas multivalent ions such as iron, calcium, magnesium, and manganese showed
retention values of 50%–90%, probably due to the association between multivalent ions and organic
material. The lignin retention at 90 ◦ C was 90%, increasing the concentration from 56 to 160 g·L−1 .
Similar results were achieved in later studies [34,35,54].
Black liquor taken directly from the pulping process has been separated at temperatures exceeding
100 ◦ C using ceramic UF membranes with MWCO values of 5 and 15 kDa [37]. Lignin retention
declined from 45% at 90 ◦ C to 22% at 145 ◦ C for the 5 kDa membrane with a TMP of 1 bar. This study
indicated that the maximum rate of temperature change should be 2–3 ◦ C·min−1 in order to prevent
membrane bursting [37]. Black liquor has also been separated using ceramic UF and NF membranes
with MWCO values of 1, 5, and 15 kDa to extract the small lignin fraction in a highly pure form [38].
The authors reported that black liquors from hardwood and softwood showed different flux behaviors,
with the hardwood liquor achieving a lower flux due to the higher average molecular weight of
the lignin fraction. After precipitation, the lignin fraction with a molecular weight less than 1 kDa
increased from 19% to 44%.
The fractionation of black liquor has also been compared using a variety of membranes made from
different materials, e.g., cellulose acetate, aluminum oxide, polyaryl ether ketone, polyacrylonitrile,
and polyethersulfone. The percentage of each lignin fraction is summarized in Table 4. Continuous
filtration was conducted for 40 days using a 0.2 µm ceramic membrane at a TMP of 2 bar and a
cross-flow velocity of 2.3 m·s−1 . The flux declined from 310 to 150 L·m−2 ·h−1 over 16 days, and
the lignin retention was 80% [39]. The filtration of black liquor with zirconium dioxide membranes
achieved a pressure-independent flux of 52 L·m−2 ·h−1 over the pressure range 3–5 bar at a cross-flow
velocity of 2.1 m·s−1 . The authors proposed that the lignin absorbed specifically at the membrane
surface and formed a gel, as confirmed by the retention of organic molecules caused by the compression
of the gel layer [40].
Table 4. Molecular weight distribution of lignin in black liquor [39].
Molecular Weight in kDa
Percentage Abundance
>60
60–30
30–10
10–6
6–3
<3
61.8
21.8
1.2
1.8
2.4
1.0
The filtration of black liquor with a rotating disk membrane module using a 5 kDa cellulose acetate
membrane required a centrifugation step due to the large number of fibers [41]. The study considered
Membranes 2016, 6, 42
7 of 13
the effect of the disk rotation speed (0–600 rpm), TMP (~3–8 bar) and stirring speed (0–1000 rpm).
An increase in the disk rotation speed from 0 to 450 rpm increased the flux by 23%, and the flux
increased by 60% after filtration for 60 min compared to a non-stirred system. The rotating membrane
induced turbulence on the feed side which reduced fouling. A 33% higher flux was achieved by
increasing the feed stirring speed from 0 to 1000 rpm [41].
More than 90% of organic contaminants were removed from black liquor using a cascade
of ultrafiltration, acid precipitation, crystallization and nanofiltration (0.4 kDa), increasing the
concentration of lignin in the retentate from 82 to 230 g·L−1 [42]. Another combination of ultrafiltration
and nanofiltration was shown to be economically viable for the production of a pure lignin fraction [44].
The UF step comprised a 20 kDa aluminum oxide/titanium oxide membrane at 90 ◦ C with a TMP
of 2.0 bar and a cross-flow velocity of 5 m·s−1 , whereas the NF step comprised a 1 kDa aluminum
oxide membrane, increasing the flux by 48%. The lignin concentration increased from 62 to 285 g·L−1 .
Pretreatment of the black liquor increased the economic value of the purified lignin by 450%, although
the use of ceramic membranes increased the process cost by 300% [44].
Remarkable differences, related to the solubility and the viscosity, were observed by comparison
of lignin obtained from UF (5 kDa) and lignin from the Lignoboost process. The filtration method
was a large scale version of the previously described system from Keyoumu et al. [38]. First of all,
both lignin types were soluble in moderately polar solvents (ethanol, acetic acid, methanol). However,
highly polar solvents like water and apolar solvents like hexane are quite poor solvents for lignin.
Nonetheless, the low molecular lignin from the UF was easier to dissolve in good lignin solvents than
the unfractionated lignin from the Lignoboost process. Additionally, the viscosity of the filtered lignin
was lower at the same concentration than lignin obtained from the Lignoboost process [45].
3.2. Membrane Technology for Treatment of Spent Sulfite Liquors
In contrast to the filtration of black liquor, relatively little progress has been made with the
development of filtration strategies for spent sulfite liquor. The recovery of lignosulfonates using
polysulfone, cellulose acetate, and fluoropolymer UF and NF membranes with MWCO values of 1, 5,
10, 20, 50, and 100 kDa has been studied in a stirred filtration cell at different TMPs (1.0–7.6 bar) with
three different dilutions of the spent liquor (neat, 1:1 dilution, and 1:5 dilution) [49]. As anticipated, the
flux increased at higher MWCO values and a wider size range of lignosulfonates passed through the
fluoropolymer membranes because the MWCO range was broader. Therefore, there was greater flux
through the fluoropolymer membranes but less retention compared to the polysulfone membranes.
The dilution experiments showed that the lignin rejection efficiency declined but the permeate flux
increased at higher dilutions. The membrane pores, blocked by low-molecular-weight lignin, were
continuously cleaned, hence the lower rejection, but rejection increased in line with the TMP due to the
effects of concentration polarization. Polysulfone membranes with high MWCO values are therefore
ideal for the recovery of lignosulfonates because of their high flux and >80% rejection [49].
Lignosulfonate fractionation by amine extraction and ultrafiltration has been compared using
polyethersulfone and cellulose UF membranes in cross-flow mode with MWCO values of 1, 10, 50, and
100 kDa. Amine extraction produced three fractions at a time and consumed large amounts of organic
solvent, whereas the UF membranes generated five fractions and thus reduced the polydispersity
within each fraction. Ultrafiltration also provided more information about the mass distribution in the
spent sulfite liquor components. However, there was no significant difference between the methods in
terms of total lignin recovery [50].
Tubular ceramic aluminum oxide/titanium oxide membranes have been tested for the treatment
of spent sulfite liquor to reduce the chemical oxygen demand as well as residual lignin levels. MF,
UF and NF membranes with pore sizes from 0.05 to 0.2 µm and MWCO values ranging from 1 kDa
to 20 kDa were tested in different single and multiplex configurations to identify the most efficient
strategy [51]. The authors concluded that a two-stage MF/UF process achieved the most promising
results, reducing the chemical oxygen demand by 45% and the concentration of the residual lignin
Membranes 2016, 6, 42
8 of 13
by 73% [51]. The ideal configuration comprised a 0.1-µm MF membrane with a crossflow velocity of
5.6 m·s−1 and a 20 kDa UF membrane with a crossflow velocity of 4.0 m·s−1 with an overall TMP
of 2 bar.
Flat cellulose acetate, regenerated cellulose, and polysulfone membranes with MWCO values of
0.5, 2, 3, and 10 kDa were used for the separation of hemicellulose and lignosulfonates from spent
sulfite liquor. The filtration was carried out in a stirred batch filtration unit under a TMP of 3.6 bar.
The flux decreased for the 3 kDa membrane from 39.6 L·m−2 ·h−1 to 8.4 L·m−2 ·h−1 during a 180 min
batch fermentation. In this time, a volume reduction of 60% was achieved. The filtration results
showed that lignosulfonate as well as hemicellulose can be recovered from all tested membranes. After
a volume reduction of 65%, a final lignin rejection of 65%, 52%, and 77% could be achieved for the 10,
3, and 2 kDa membrane, respectively. The 3 kDa membrane, made of regenerated cellulose, gave the
best results for ligninsulfonate recovery with a high rejection and selectivity of lignosulfonate [52].
Flat UF and NF membranes made of regenerated cellulose have been investigated for the
pretreatment of spent sulfite liquor for the bio-based succinic acid production with Actinobacillus succinogenes
and Basfia succiniciproducens. In this study, membranes with MWCO values of 10, 5, 3, and 0.5 kDa
were used. The UF experiments were carried out in a stirred ultrafiltration cell at a TMP of 3.0 bar. Per
batch, 400 mL of 10 times diluted spent sulfite liquor was filtered. Thereby a flux of 35.7 L·m−2 ·h−1
was achieved for the 10 kDa. The lignosulfonate concentration in the permeate increased from
~37 g·L−1 to 104.3, 119.2, 157.1 g·L−1 for the 10, 5, and 3 kDa membranes, respectively. The NF
with the 0.5 kDa membrane was conducted with a vibratory share-enhanced filtration unit that used
oscillatory vibrations to induce high shear rates on the membrane surface whereby membrane fouling
was reduced. The seven times diluted spent sulfide liquor had a temperature of 54–75 ◦ C during the
filtration. The TMP was 20.8 bar at the beginning of the filtration and was gradually increased to 31 bar
at the end of the filtration. The flux decreased from 44.0 L·m−2 ·h−1 to 10.7 L·m−2 ·h−1 in the course of
the filtration. The lignosulfonate concentration increased from the initial concentration of 60.0 g·L−1 to
166.1 g·L−1 in the retentate. The 0.5 kDa NF membrane achieved a lignosulfonate recovery of 95.6%.
The fermentation experiments with the permeate from the NF resulted in a production of ~52 g·L−1
succinic acid whereas the use of the permeate from the 3 kDa UF membrane resulted in 71.8 g·L−1
succinic acid [53].
4. Membrane Fouling and Cleaning Strategies
Membrane fouling is defined as the irreversible formation of deposits on the active surface of a
membrane, causing a decline in flux and a loss of performance. This is a particular problem during
the filtration of brackish water, seawater, and industrial wastewater [54]. The filtration of black liquor
and spent sulfite liquor is prone to fouling and this becomes a significant cost factor in the context of
industrial-scale processes. Therefore, effective cleaning strategies are necessary to reduce the rate of
fouling and prolong the life of membranes.
One potential solution involves a combination of rinsing and chemical cleaning [33].
The membrane was rinsed with permeate collected during the filtration because this has the same
pH as the original feed solution and potential foulants are, therefore, likely to dissolve. Rinsing was
carried out at the operational temperature and at a TMP of 0.5 bar. Chemical cleaning was carried
out with 0.25 wt % Ultrasil 11 (Henkel, Germany) at 60 ◦ C and a TMP of 0.5 bar until the initial pure
water flux was restored. Ultrasil 11 is an alkaline cleaning agent free of ethylenediaminetetraacetic
acid (EDTA) and nitrilotriacetic acid (NTA). The initial pure water flux of the ceramic membrane
(MWCO = 15 kDa) was completely recovered [33]. Rinsing alone was shown to restore 70%–80% of the
initial pure water flux, whereas the combination of rinsing and chemical cleaning restored 90%–95% of
the initial value [8].
A four-step process has been developed for inorganic membranes: 30 min washing with tap water,
30 min washing with 0.2 mol·L−1 sodium hydroxide, 30 min in the furnace at 550 ◦ C, and 30 min
washing with 0.2 mol·L−1 hydrochloric acid. The initial pure water flux was completely recovered [39].
Membranes 2016, 6, 42
9 of 13
For organic membranes, the same procedure can be used without the furnace step, restoring 50%–80%
of the initial pure water flux [39].
The cleaning of ceramic membranes after the filtration of spent sulfite liquor was carried out using
1% sodium hydroxide at 60 ◦ C for 1 h followed by rinsing with distilled water. For a 20 kDa membrane,
the
pure 2016,
water
flux was restored to 98% of the initial value, and for a 1 kDa membrane cleaning
Membranes
6, 42
9 of 12
restored 80% of the pure water flux [51]. A back-flush for 10 s at a pressure of 4 bar was also shown
flushes.
the but
interval
was was
60 min
theefficient
averagewhen
flux increased
16%, but
with an
intervalflushes.
of 120
to
reduceWhen
fouling,
cleaning
more
there wasby
a longer
interval
between
min the
48% the
[51].
Our experiments
forby
the
treatment
ofan
spent
sulfite
liquor
When
theaverage
intervalflux
was by
60 min
average
flux increased
16%,
but with
interval
of 120
minalso
the
focused flux
on chemical
cleaning
strategies for
membranes.
Using
anliquor
aluminum
oxide UF
average
by 48% [51].
Our experiments
forceramic
the treatment
of spent
sulfite
also focused
on
membrane
with a MWCO
value
of 5 kDamembranes.
manufactured
by atech
innovations
(Gladbeck,
Germany),
chemical
cleaning
strategies
for ceramic
Using
an aluminum
oxide
UF membrane
with
initial value
clean water
flux
was completely
recovered
after cleaning
for 4Germany),
h with Ultrasil
11 at 60
°C
athe
MWCO
of 5 kDa
manufactured
by atech
innovations
(Gladbeck,
the initial
clean
and a TMP
of 0.5
bar. The permeate
over
a filtration
lasting
17 h 11
is shown
inand
Figure
2. The
water
flux was
completely
recoveredflux
after
cleaning
for 4 run
h with
Ultrasil
at 60 ◦ C
a TMP
of
permeate
flux
declined
after
3 ah filtration
by approximately
28.0 in
to Figure
11.8 L∙m
∙h−1.permeate
The average
0.5
bar. The
permeate
flux
over
run lasting60%
17 hfrom
is shown
2. −2
The
flux
−2 ·hfiltration
−1 . The average
permeateafter
flux 3was
12.5
L∙m−2∙h−1 The 60%
cleanfrom
water
fluxes
before
of spent permeate
sulfite liquor,
declined
h by
approximately
28.0
to 11.8
L·mthe
flux
−
2
−
1
after12.5
filtration,
cleaning
compared
in Table
5 and sulfite
are shown
in Figure
2. After
was
L·m and
·h after
The chemical
clean water
fluxes are
before
the filtration
of spent
liquor,
after filtration,
the treatment
of spent
sulfiteare
liquor,
the membrane
physically
cleaned
with2.pure
water,
restoring
and
after chemical
cleaning
compared
in Table was
5 and
are shown
in Figure
After
the treatment
56%
of the
initial
pure the
water
flux, andwas
chemical
cleaning
restored
remaining
44%. The56%
flux-time
of
spent
sulfite
liquor,
membrane
physically
cleaned
withthe
pure
water, restoring
of the
curve of
thewater
filtration
the typical
profile,restored
comprising
rapid initial
decline
(0–50 min)curve
caused
initial
pure
flux,shows
and chemical
cleaning
the aremaining
44%.
The flux-time
of
by fast
pore shows
blocking
by thecomprising
formation aand
growth
a cake
layer,
adsorption,
the
filtration
the followed
typical profile,
rapid
initial of
decline
(0–50
min)
caused by and
fast
concentration
polarization
(50–1000
min).
flux, layer,
reflecting
an equilibrium
between
pore
blocking followed
by the
formation
andSteady-state
growth of a cake
adsorption,
and concentration
membrane and
cake layer
at the
cross an
flow
velocity, and
temperature
shown,
was
polarization
(50–1000
min).resistance
Steady-state
flux,TMP,
reflecting
equilibrium
between
membrane
and cake
not accomplished
in TMP,
this experiment
because and
the temperature
filtration was
terminated
after
1000 h [55].
layer
resistance at the
cross flow velocity,
shown,
was not
accomplished
in
Irreversible
fouling
was not
observedwas
afterterminated
the first filtration
cycle.
Chemical
cleaning
restored
the
this
experiment
because
the filtration
after 1000
h [55].
Irreversible
fouling
was not
initial water
flux
therefore,
complex cleaning
and labor-intensive
strategies
to recover
observed
after
thecompletely,
first filtration
cycle. Chemical
restored thecleaning
initial water
flux completely,
the initial complex
flux was and
not necessary.
therefore,
labor-intensive cleaning strategies to recover the initial flux was not necessary.
Figure 2.
2. Filtration
Filtration of
of spent
spent sulfite
sulfite liquor
liquor using
using aa 55 kDa
kDa ceramic
ceramic membrane
membrane at
at 60
60 ◦°C,
with an
an average
average
Figure
C, with
transmembrane
pressure
(TMP)
of
0.4
bar.
The
blue
dots
represent
the
initial
pure
water
flux,
the
transmembrane pressure (TMP) of 0.4 bar. The blue dots represent the initial pure water flux, the water
water
flux
after
the
filtration,
and
the
water
flux
after
the
chemical
cleaning.
flux after the filtration, and the water flux after the chemical cleaning.
Table 5. Water fluxes at different TMPs before the filtration of spent sulfite liquor, after filtration, and
after chemical cleaning.
Pressure (bar)
0.5
1.0
1.5
Water Flux before
Filtration
(L∙m−2∙h−1)
45.5
88.7
140.7
Water Flux after Filtration
before Chemical Cleaning
(L∙m−2∙h−1)
24.5
51.4
79.0
Water Flux after
Chemical Cleaning
(L∙m−2∙h−1)
42.9
94.1
145.3
Membranes 2016, 6, 42
10 of 13
Table 5. Water fluxes at different TMPs before the filtration of spent sulfite liquor, after filtration, and
after chemical cleaning.
Pressure (bar)
Water Flux before
Filtration (L·m−2 ·h−1 )
Water Flux after Filtration
before Chemical Cleaning
(L·m−2 ·h−1 )
Water Flux after
Chemical Cleaning
(L·m−2 ·h−1 )
0.5
1.0
1.5
2.0
45.5
88.7
140.7
197.5
24.5
51.4
79.0
110.6
42.9
94.1
145.3
193.0
5. Summary
Every pulping process in the world produces vast amounts of wastewater containing modified
lignin, which is usually incinerated to provide energy despite the potential to recover the lignin and
use it for the manufacture of valuable chemical products. Membrane technology offers an economical
and environmentally beneficial platform for the recovery of lignin. Organic and inorganic membranes
with MWCO values of 0.4–100 kDa have been investigated, using diverse methods such as radial,
rectangular cross-flow, and dead-end filtration combined with stirred rotating disc modules and
plate-and-frame modules. The impact of different filtration parameters such as TMP, feed temperature,
cross-flow velocity, and the Reynolds number have been studied in detail, resulting in processes that
achieve lignin concentrations of up to 285 g·L−1 and purities of up to 78% [8,44]. Membrane filtration
combined with diafiltration also reduced the ash content of kraft liquor from 4.7% to 2.7% [8].
Membrane cleaning strategies have been developed to prevent fouling and restore membrane
flux after lignin purification. A combination of rinsing (with permeate) and cleaning with Ultrasil 11
achieved the best results, and regular back-flushing was shown to prevent fouling and increase the
average flux by up to 48% [51].
Economic models suggest that membrane-based recovery can produce pure lignin at a cost of
€46–120 per tonne, although the inclusion of a diafiltration step would cost more [43]. Therefore, the
membrane-based filtration of wastewater from the pulp and paper industry holds great promise for
the economical and environmentally-sustainable recovery of concentrated and highly pure lignin, with
strategies in place to maximize membrane performance and longevity. Membrane technology could
face strong competition from the Lignoboost process, but the latter is only suitable for black liquor
and is also more difficult to scale-up. Therefore, membrane technology offers a unique combination
of versatility, scalability, and economy that makes it suitable for the recovery of lignin from diverse
industrial processes.
Author Contributions: Daniel Humpert wrote the paper. Mehrdad Ebrahimi and Peter Czermak revised the
manuscript. Peter Czermak is the head of the research program.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
Pabby, A.; Rizvi, S.; Sastre, A.M. Handbook of Membrane Separations; CRC Press Taylor & Francis Group:
New York, NY, USA, 2009; Volume 53.
Boerjan, W.; Ralph, J.; Baucher, M. Lignin Biosynthesis. Annu. Rev. Plant Biol. 2003, 54, 519–546. [CrossRef]
[PubMed]
Stephen Davidson, R.; Dunn, L.A.; Castellan, A.; Nourmamode, A. A study of the photobleaching and
photoyellowing of paper containing lignin using fluorescence spectroscopy. J. Photochem. Photobiol. A Chem.
1991, 58, 349–359. [CrossRef]
Bajpai, P. Biotechnology for Pulp and Paper Processing; Intergovernmental Panel on Climate Change, Ed.;
Springer US: Boston, MA, USA, 2012; Volume 1.
Membranes 2016, 6, 42
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
11 of 13
Busch, R.; Hirth, T.; Liese, A.; Nordhoff, S.; Puls, J.; Pulz, O.; Sell, D.; Syldatk, C.; Ulber, R. Nutzung
nachwachsender Rohstoffe in der industriellen Stoffproduktion. Chemie Ing. Tech. 2006, 78, 219–228.
[CrossRef]
Žabková, M.; da Silva, E.A.B.; Rodrigues, A.E. Recovery of vanillin from lignin/vanillin mixture by using
tubular ceramic ultrafiltration membranes. J. Memb. Sci. 2007, 301, 221–237. [CrossRef]
Pye, E.K. Industrial Lignin Production and Applications. In Biorefineries-Industrial Processes and Products;
Wiley-VCH Verlag GmbH: Weinheim, Germany, 2006; Volume 2, pp. 165–200.
Wallberg, O.; Jönsson, A.-S.; Wimmerstedt, R. Fractionation and concentration of kraft black liquor lignin
with ultrafiltration. Desalination 2003, 154, 187–199. [CrossRef]
Biermann, C.J. Handbook of pulping and papermaking; Elsevier Inc.: Amsterdam, The Netherlands, 1996.
Karam, J.; Nicell, J.A. Review Potential Applications of Enzymes in Waste Treatment. J. Chem. Tech. Biotechnol.
1997, 69, 141–153. [CrossRef]
Tian, Z.; Zong, L.; Niu, R.; Wang, X.; Li, Y.; Ai, S. Recovery and characterization of lignin from alkaline straw
pulping black liquor: As feedstock for bio-oil research. J. Appl. Polym. Sci. 2015, 132. [CrossRef]
El Mansouri, N.-E.; Salvadó, J. Structural characterization of technical lignins for the production of
adhesives: Application to lignosulfonate, kraft, soda-anthraquinone, organosolv and ethanol process lignins.
Ind. Crops Prod. 2006, 24, 8–16. [CrossRef]
Argyropoulos, D.S.; Sadeghifar, H.; Cui, C.; Sen, S. Synthesis and Characterization of Poly(arylene ether
sulfone) Kraft Lignin Heat Stable Copolymers. ACS Sustain. Chem. Eng. 2014, 2, 264–271. [CrossRef]
Dizhbite, T.; Telysheva, G.; Jurkjane, V.; Viesturs, U. Characterization of the radical scavenging activity of
lignins-natural antioxidants. Bioresour. Technol. 2004, 95, 309–317. [CrossRef] [PubMed]
Pakkanen, H.; Alén, R. Molecular Mass Distribution of Lignin from the Alkaline Pulping of Hardwood,
Softwood, and Wheat Straw. J. Wood Chem. Technol. 2012, 32, 279–293. [CrossRef]
Asikkala, J.; Tamminen, T.; Argyropoulos, D.S. Accurate and Reproducible Determination of Lignin Molar
Mass by Acetobromination. J. Agric. Food Chem. 2012, 60, 8968–8973. [CrossRef] [PubMed]
SUN, R.; TOMKINSON, J.; BOLTON, J. Separation and Characterization of Lignins from the Black Liquor of
Oil Palm Trunk Fiber Pulping. Sep. Sci. Technol. 1999, 34, 3045–3058. [CrossRef]
Olivares, M.; Guzmán, J.A.; Natho, A.; Saavedra, A. Kraft lignin utilization in adhesives. Wood Sci. Technol.
1988, 22, 157–165. [CrossRef]
Xavier, A.M.R.B.; Correia, M.F.; Pereira, S.R.; Evtuguin, D.V. Second-generation bioethanol from eucalypt
sulphite spent liquor. Bioresour. Technol. 2010, 101, 2755–2761. [CrossRef] [PubMed]
Novy, V.; Krahulec, S.; Longus, K.; Klimacek, M.; Nidetzky, B. Co-fermentation of hexose and pentose sugars
in a spent sulfite liquor matrix with genetically modified Saccharomyces cerevisiae. Bioresour. Technol. 2013,
130, 439–448. [CrossRef] [PubMed]
Marques, A.P.; Evtuguin, D.V.; Magina, S.; Amado, F.M.L.; Prates, A. Chemical Composition of Spent Liquors
from Acidic Magnesium–Based Sulphite Pulping of Eucalyptus globulus. J. Wood Chem. Technol. 2009, 29,
322–336. [CrossRef]
Weis, A.; Michael, B.; Marianne, N. The chemical cleaning of polymeric UF membranes fouled with spent
sulphite liquor over multiple operational cycles. J. Memb. Sci. 2003, 216, 67–79. [CrossRef]
Rueda, C.; Calvo, P.A.; Moncalián, G.; Ruiz, G.; Coz, A. Biorefinery options to valorize the spent liquor from
sulfite pulping. J. Chem. Technol. Biotechnol. 2015, 90, 2218–2226. [CrossRef]
Nigam, J.N. Ethanol production from hardwood spent sulfite liquor using an adapted strain of Pichia stipitis.
J. Ind. Microbiol. Biotechnol. 2001, 26, 145–150. [CrossRef] [PubMed]
Yu, S.; Wayman, M.; Parekh, S.K. Fermentation to ethanol of pentose-containing spent sulphite liquor.
Biotechnol. Bioeng. 1987, 29, 1144–1150. [CrossRef] [PubMed]
Ringena, O.; Lebioda, S.; Lehnen, R.; Saake, B. Size-exclusion chromatography of technical lignins in dimethyl
sulfoxide/water and dimethylacetamide. J. Chromatogr. A 2006, 1102, 154–163. [CrossRef] [PubMed]
Area, M.C.; Felissia, F.E.; Venica, A.; Valade, J.L. Upgrading spent liquors from NSSC process. II.
Lignosulfonates and hemicelluloses molecular weight distributions. Cellul. Chem. Technol. 2000, 34, 525–535.
Kouisni, L.; Fang, Y.; Paleologou, M.; Ahvazi, B.; Hawari, J.; Zhang, Y.; Wang, X.-M. Kraft lignin recovery
and its use in the preparation of lignin-based phenol formaldehyde resins for plywood. Cellul. Chem. Technol.
2011, 45, 515–520.
Membranes 2016, 6, 42
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
12 of 13
Zhu, W. Precipitation of Kraft Lignin Yield and Equilibrium; Chalmers University of Technology: Göteborg,
Sweden, 2015.
Uloth, V.; Wearing, J.T. Kraft Lignin Recovery: Acid Precipitation Versus Ultrafiltration, Part II: Technology
and economics. Pulp Paper Can. 1989, 90, T357–T360.
Uloth, V. Kraft Lignin Recovery: Acid Precipitation Versus Ultrafiltration, Part I: Laboratory Test Results.
Pulp Paper Can. 1989, 90, T310–T314.
Satyanarayana, S.V.; Bhattacharya, P.K.; De, S. Flux decline during ultrafiltration of kraft black liquor using
different flow modules: A comparative study. Sep. Purif. Technol. 2000, 20, 155–167. [CrossRef]
Wallberg, O.; Jönsson, A.-S.; Wimmerstedt, R. Ultrafiltration of kraft black liquor with a ceramic membrane.
Desalination 2003, 156, 145–153. [CrossRef]
Jönsson, A.-S.; Nordin, A.-K.; Wallberg, O. Concentration and purification of lignin in hardwood kraft
pulping liquor by ultrafiltration and nanofiltration. Chem. Eng. Res. Des. 2008, 86, 1271–1280. [CrossRef]
Holmqvist, A; Wallberg, O.; Jönsson, A.-S. Ultrafiltration of Kraft Black Liquor from Two Swedish Pulp Mills.
Chem. Eng. Res. Des. 2005, 83, 994–999. [CrossRef]
De, S.; Bhattacharya, P.K. Flux prediction of black liquor in cross flow ultrafiltration using low and high
rejecting membranes. J. Memb. Sci. 1996, 109, 109–123. [CrossRef]
Wallberg, O.; Jönsson, A.-S. Separation of lignin in kraft cooking liquor from a continuous digester by
ultrafiltration at temperatures above 100 ◦ C. Desalination 2006, 195, 187–200. [CrossRef]
Keyoumu, A.; Sjödahl, R.; Henriksson, G.; Ek, M.; Gellerstedt, G.; Lindström, M.E. Continuous nano- and
ultra-filtration of kraft pulping black liquor with ceramic filters. Ind. Crops Prod. 2004, 20, 143–150. [CrossRef]
Liu, G.; Liu, Y.; Ni, J.; Shi, H.; Qian, Y. Treatability of kraft spent liquor by microfiltration and ultrafiltration.
Desalination 2004, 160, 131–141. [CrossRef]
Dafinov, A.; Font, J.; Garcia-Valls, R. Processing of black liquors by UF/NF ceramic membranes. Desalination
2005, 173, 83–90. [CrossRef]
Bhattacharjee, S.; Datta, S.; Bhattacharjee, C. Performance study during ultrafiltration of Kraft black liquor
using rotating disk membrane module. J. Clean. Prod. 2006, 14, 497–504. [CrossRef]
Niemi, H.; Lahti, J.; Hatakka, H.; Kärki, S.; Rovio, S.; Kallioinen, M.; Mänttäri, M.; Louhi-Kultanen, M.
Fractionation of Organic and Inorganic Compounds from Black Liquor by Combining Membrane Separation
and Crystallization. Chem. Eng. Technol. 2011, 34, 593–598. [CrossRef]
Tomani, P. The lignoboost process. Cellul. Chem. Technol. 2010, 44, 53–58.
Arkell, A.; Olsson, J.; Wallberg, O. Process performance in lignin separation from softwood black liquor by
membrane filtration. Chem. Eng. Res. Des. 2014, 92, 1792–1800. [CrossRef]
Giummarella, N.; Lindgren, C.; Lindström, M.; Henriksso, G. Lignin Prepared by Ultrafiltration of Black
Liquor: Investigation of Solubility, Viscosity, and Ash Content. BioResources 2016, 11, 3494–3510. [CrossRef]
Toledano, A.; Serrano, L.; Garcia, A.; Mondragon, I.; Labidi, J. Comparative study of lignin fractionation by
ultrafiltration and selective precipitation. Chem. Eng. J. 2010, 157, 93–99. [CrossRef]
Padilla, V.; Rangel, M.G.; Bullon, J.; Rodríguez-Malaver, A.; Gonzalez, A.M.; Rojas, O.J. Surface Activity of
Lignin Fractions Obtained By Membrane-Separation Technologies of Industrial Black Liquors. Iberoam. Congr.
Pulp Pap. Res. 2002, 10, 10–12.
Sevastyanova, O.; Helander, M.; Chowdhury, S.; Lange, H.; Wedin, H.; Zhang, L.; Ek, M.; Kadla, J.F.;
Crestini, C.; Lindström, M.E. Tailoring the molecular and thermo-mechanical properties of kraft lignin by
ultrafiltration. J. Appl. Polym. Sci. 2014, 131, 40799(1)–40799(11). [CrossRef]
Bhattacharya, P.K.; Todi, R.K.; Tiwari, M.; Bhattacharjee, C.; Bhattacharjee, S.; Datta, S. Studies on
ultrafiltration of spent sulfite liquor using various membranes for the recovery of lignosulphonates.
Desalination 2005, 174, 287–297. [CrossRef]
Ringena, O.; Saake, B.; Lehnen, R. Isolation and fractionation of lignosulfonates by amine extraction and
ultrafiltration: A comparative study. Holzforschung 2005, 59, 405–412. [CrossRef]
Ebrahimi, M.; Busse, N.; Kerker, S.; Schmitz, O.; Hilpert, M.; Czermak, P. Treatment of the Bleaching Effluent
from Sulfite Pulp Production by Ceramic Membrane Filtration. Membranes. 2015, 6. [CrossRef] [PubMed]
Area, M.C.; Martos, M.S.; Felissia, F.E.; Venica, A.D.; Valade, J.L. Upgrading Spent Liquors From Nssc Process
III. Separation of Spent Liquors Components By Ultrafiltration. Cellul. Chem. Technol. 2000, 34, 173–182.
Membranes 2016, 6, 42
53.
54.
55.
13 of 13
Pateraki, C.; Ladakis, D.; Stragier, L.; Verstraete, W.; Kookos, I.; Seraphim, P.; Koutinas, A. Pretreatment of
spent sulphite liquor via ultrafiltration and nanofiltration for bio-based succinic acid production. J. Biotechnol.
2016, 233, 95–105. [CrossRef] [PubMed]
Peña, N.; Gallego, S.; del Vigo, F.; Chesters, S.P. Evaluating impact of fouling on reverse osmosis membranes
performance. Desalin. Water Treat. 2013, 51, 958–968. [CrossRef]
Song, L. Flux decline in crossflow microfiltration and ultrafiltration: Mechanisms and modeling of membrane
fouling. J. Memb. Sci. 1998, 139, 183–200. [CrossRef]
© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).